The linkage between cryptocurrencies and macro-financial parameters : a data mining approach

Digital currencies have increased their effectiveness in recent years and have started to see significant demand in international markets. Bitcoin stands out from the other cryptocurrencies in considering the transaction volume and the rate of return. In this study, Bitcoin is estimated by using a decision tree method which is among the data mining methodology. The variables used in the decision tree created in the estimation of Bitcoin are the S&P 500 stock index, gold prices, oil prices, Euro/Dollar exchange rate, and FED Treasury bill interest rate. When the experimental results were examined, it was observed that the decision tree C4.5 algorithm was an appropriate method with the correct classification percentage of 73% in estimating the direction of Bitcoin. Also, the results obtained from the decision tree show that Bitcoin is related to S&P 500 index among macro-financial indicators similar to the results of econometric models used in the literature.